Abstract Cloud resource allocation faces critical challenges in balancing economic efficiency with technical constraints in dynamic environments. Traditional approaches often result in suboptimal trade-offs, marked by low resource utilization and elevated costs. The proposed research creates a new hybrid system that combines combinatorial double auction algorithms with a genetic algorithm (GA) to optimize cloud resource allocation. The proposed system uses adaptive mutation and crossover operations for optimization. It also applies a hybrid fitness function that minimizes costs and distributes the datacenter load equally. A three-tier architecture facilitates seamless integration with simulation tools for real-time validation and feedback. The proposed framework shows superior performance in experimental tests, which leads to significant cost savings, increases resource use, and maintains stable prices above standard approaches. This paper proposes a genetic algorithm–driven combinatorial double auction framework for efficient cloud resource allocation. The proposed approach introduces a feasibility-aware chromosome encoding and a dual-objective fitness function that jointly optimizes provisioning cost and load balance under capacity constraints. Extensive CloudSim-based simulations demonstrate that the proposed method achieves improved resource utilization, enhanced fairness, and stable convergence compared to existing auction-based and heuristic approaches. These results highlight the effectiveness of the proposed framework for scalable and dynamic cloud environments. Experimental results show that the proposed approach reduces total allocation cost by up to 34.8%, achieves a fairness index of 0.798 with approximately 93% resource utilization, and scales efficiently with an average execution time of 7.16 seconds in large-scale simulated cloud environments.
Zeshan Iqbal (Mon,) studied this question.